1.5 vs. 2x - build problem
Geoff Gerrietts
geoff at gerrietts.net
Thu Sep 11 13:11:42 EDT 2003
More information about the Python-list mailing list
Thu Sep 11 13:11:42 EDT 2003
- Previous message (by thread): 1.5 vs. 2x - build problem
- Next message (by thread): How to compile 3rd party python into single large binary
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Quoting Fritz Wuehler (fritz at expires-200309.rodent.frell.eu.org): > I downloaded a tarball which doesn't compile. The error message > says I need python 2x but all I have is 1.5. Can I fix the > compatability problem and get it to work on 1.5? My experience has been that most (and not all!) of the applications that require Python 2.x, do it mostly for the sake of convenience. This means that they'll use func(*args, **kw) instead of apply(funcs, args, kw) or maybe mylist = [str(x) for x in yourlist] instead of mylist = map(str, yourlist) and x += 1 instead of x = x + 1 There are other features you'll find that don't work -- Python has gotten a lot more convenient since 1.5.2! -- but most of the time, the 2.x dependencies are convenience features. Some features are more difficult to port back than others, though. If you're dealing with something that (for instance) has unicode requirements, you're just not gonna be able to do what you need on 1.5.2. So the answer is: probably; I have only run across one app so far that I couldn't eventually port back to 1.5.2 at least well enough that it was running for me. That app was Pyana. It may or may not be easy. Quoting Gerhard Häring (gh at ghaering.de): > On the other hand, you could invest the time and bring your > software/system up-to-date with Python 2.3. You can just install it in > /usr/local and leave your previously installed Python alone. > > FWIW I, for one, would reject any patches that'd add Python < 2.1 > compatibilty to my software. I wouldn't want to take one of my personal projects back to 1.5.2 and lose the additional expressive power that recent Pythons have given me, either. On the other hand, in my professional capacity, I have spent an assembled couple of months trying to port 200k lines of application code forward onto new versions of Python and Zope. It's going to take a couple weeks of QA and deployment work to get it all into place. On the other hand, it usually takes me an afternoon to backport someone else's extension code well enough that I can use it. Sometimes people aren't interested in the patches, sometimes they are. Sometimes I fail to produce them after making the changes. For my part, I feel bad for the next guy down the line in at least two of those cases. Many people have this cavalier "well just upgrade, wtf is wrong with you?" attitude, but it's often not as trivial as everyone would like to pretend. (Of course, sometimes it /is/ trivial, and if it is, you should do it.) --G. -- Geoff Gerrietts <geoff at gerrietts dot net> "Ordinarily he was insane, but he had lucid moments when he was merely stupid." --Heinrich Heine
- Previous message (by thread): 1.5 vs. 2x - build problem
- Next message (by thread): How to compile 3rd party python into single large binary
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Python-list mailing list